wafer-ai/gpu-perf-engineering-resources — GitHub trend

◂ BACK TO LEADERBOARD

FINDING #270 · UNIT ID 1132444278

wafer-ai/gpu-perf-engineering-resources

A curated resource list for learning AI performance engineering, from GPU fundamentals to production inference.

[ PYTHON ][ ORG ][ GITHUB ↗ ]

SURPRISE SCORE

0.00

Score Breakdown

SURPRISE0.00376

ENGAGEMENT0.42

FRESHNESS1.21

SCORE = SURPRISE × ENGAGEMENT^0.7 × FRESHNESS × VISIBILITY × CONFIDENCE

SURPRISE = WINDOW STARS / DAYS / (AUDIENCE + FLOOR)

4% OF STARS IN ARCHIVE

Growth Telemetry

VELOCITY /D

4.83

ACCEL

-0.74

RETENTION

1.9%

PEAK 2026-09-12 · FORK-RETENTION 20.0% · 145 STARS / WINDOW

Author Audience

AUDIENCE

1,247

FOLLOWERS

211

OWNER ★

4,125

Engagement Signals

FORKS

380

ISSUE AUTH

PR AUTH

UNIQUE STARGAZERS 145 / 145 (DIVERSITY 1.00)

Why This Is A Finding

wafer-ai/gpu-perf-engineering-resources собрал 145 звёзд за окно, тогда как у автора всего 211 подписчиков — эффективная аудитория ≈ 1,247. Это даёт surprise-индекс 0.00376 (звёзды относительно охвата автора, а не в абсолюте). Удержание форков 20.0% и 3 внешних контрибьюторов отделяют реальный инструмент от разовой вспышки. Акселерация отрицательная — внимание остывает после пика.

Related Findings

RANKS ABOVE 74% OF 1039 FINDINGS

#02fruitflydev/flycoinrh[ PYTHON ]0.0358 #04perseval-BLR/NeuralScreen[ PYTHON ]0.0317 #07jordan-gibbs/hyperresearch[ PYTHON ]0.0287 #08ithtelab/workbuddy-manager[ PYTHON ]0.0277 #20ayghri/i-have-adhd[ PYTHON ]0.0231

METRICS IN CONTEXT

MEDIAN ACROSS ALL 1039 FINDINGS · Δ vs MEDIAN · PERCENTILE = SHARE RANKED BELOW

METRICVALUEMEDIANΔ MEDPERCENTILE

SCORE0.000.00+0.00ABOVE 74%

VELOCITY4.832.77+2.07ABOVE 73%

RETENTION1.9%9.5%-7.5 PPABOVE 9%

FORKS3801,250-870ABOVE 30%

SURPRISE0.000.00+0.00ABOVE 77%